What's the best overview of common Micromorts?

post by Raemon · 2020-09-03T02:39:42.002Z · LW · GW · 7 comments

This is a question post.

I want to get generally oriented on how various common risks compare against each other. I've seen some of this come up in recent Covid discussion, but I'm interested in a good article that's like "Here's all the most dangerous stuff it's likely that you do, and here's how it breaks down for various sub-activities."

This question triggered by "the first few google results not being that good."

Answers

answer by Davidmanheim · 2020-09-03T05:49:34.223Z · LW(p) · GW(p)

Scott's answer is a good one - you should read "The Norm Chronicles." But I think the question has a problem. Micromorts are a time-agnostic measure of dying, and the problem is that most risks you take don't actually translate well into micromorts.

Smoking a cigarette, which reduces your life expectancy by around 30 seconds, translates into either zero micromorts, or one, depending on how you set up the question. Increasing your relative risk of dying from cancer in 30 years isn't really the same as playing Russian roulette with a 1-million-chamber gun. Similarly, a healthy 25 year old getting COVID has about a 70-micromort risk based on direct mortality from COVID. But that number ignores the risks of chronic fatigue, later complications, or reduced life expectancy (all of which we simply don't know enough to quantify well.)

The answer that health economists have to this question is the QALY, which has its own drawbacks. For example, QALYs can't uniformly measure the risks of Russian roulette, since the risk depends on the age and quality of life of the player.

What we're left with is that the actual question we want answered has a couple more dimensions than a single metric can capture - and as I have mentioned once or twice elsewhere, reductive. metrics. have various. problems.

comment by Raemon · 2020-09-03T19:38:00.776Z · LW(p) · GW(p)

I guess the actual real question is "Is there a good blogpost somewhere that 80/20s the overall question of 'what are some reasonable/least-bad-ways of quantifying risk, and gives a rough rundown of how various common activities compare'?"

It might be that another way of reframing the question is "Could someone write a LW style book review of the Norm Chronicles?" (looks like I should perhaps read it in full, but it'd still be useful to have the key takeaways summarized)

Replies from: Davidmanheim
comment by Davidmanheim · 2020-09-04T08:57:32.990Z · LW(p) · GW(p)
"Could someone write a LW style book review of the Norm Chronicles?"

Endorsed.

answer by alex_lw · 2020-09-04T05:00:30.764Z · LW(p) · GW(p)

I'm pretty sure this chart has been posted here before in a similar context, but probably won't hurt reposting.

comment by Raemon · 2020-09-04T17:42:29.584Z · LW(p) · GW(p)

This actually looks pretty close to the thing I wanted, thanks!

comment by Vitor · 2020-09-06T11:12:39.212Z · LW(p) · GW(p)

This chart might be misleading in that it doesn't account for the impact of a person's skill on the danger. Some of these activities have a fixed risk (commercial flying), while others directly depend on how fit/agile/careful the person is, so the risk probably varies by orders of magnitude between individuals (motorcycling). At the more dangerous end, I'd expect the risk to be underestimated significantly: many people go skiing, only a few very fit people attempt to climb Everest.

7 comments

Comments sorted by top scores.